Re: How to write jobs in postgresql

From: "Jim C(dot) Nasby" <jnasby(at)pervasive(dot)com>
To: Guy Rouillier <guyr(at)masergy(dot)com>
Cc: Roman Neuhauser <neuhauser(at)sigpipe(dot)cz>, pgsql-general(at)postgresql(dot)org
Subject: Re: How to write jobs in postgresql
Date: 2005-09-06 22:19:58
Message-ID: 20050906221958.GZ60481@pervasive.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Tue, Sep 06, 2005 at 04:29:31PM -0500, Guy Rouillier wrote:
> Roman Neuhauser wrote:
> > # jnasby(at)pervasive(dot)com / 2005-08-15 20:25:20 -0500:
> >> On Tue, Aug 09, 2005 at 03:26:27PM -0500, Guy Rouillier wrote:
> >>> chiranjeevi.i wrote:
> >>>> Hi Team Members,
> >>>>
> >>>> Is it possible to write jobs in postgresql & if possible how
> >>>> should I write .please help me.
> >>>
> >>> See pgjob in pgfoundry: http://pgfoundry.org/projects/pgjob/. It's
> >>> in the planning stages.
> >>
> >> Actually, it's currently in the going nowhere stage since no one's
> >> expressed any interest in it. Anyone who's interested is encouraged
> >> to join the mailing list and post what they'd like to see from the
> >> project.
> >
> > What's the advantage over system-native (cron etc) means?
>
> Search the archives, you'll find numerous discussions on this topic,
> including the one that prompted Jim to create the project. As of now,
> the project is pre-concept stage, making it impossible to identify its
> advantages. One possible advantage would be recording job schedules in
> the database where they can be easily managed, but that's small. A
> bigger advantage can be seen in the approach that Oracle takes, where
> authentication happens when the job is created. So you don't need to
> provide credentials at run time, which in the case of cron jobs would
> mean putting passwords into shell scripts.

As Guy points out, this is all in a very formative stage right now
(although someone is supposed to be sending me some code), but here's
some other advantages:

This would be platform-independant, which is important now that we
support windows natively.

The interface would be in SQL (probably a set of functions), making it
much easier to control programatically.

Scheduling modes that are either difficult or impossible to do with cron
become available, such as sub-minute scheduling (ie: every 30 seconds),
running something at server start-up/shut-down, running something based
on a notify, etc.

I encourage anyone who's interested in this to join the mailing list at
http://lists.pgfoundry.org/mailman/listinfo/pgjob-devel
--
Jim C. Nasby, Sr. Engineering Consultant jnasby(at)pervasive(dot)com
Pervasive Software http://pervasive.com work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Ben 2005-09-06 22:26:29 Re: Basic locking question
Previous Message Joshua D. Drake 2005-09-06 22:13:58 Re: Basic locking question